Protocol Analysis

The Fifth Protocol

Electronification of blocks in the bond market has been the holy grail. Four established protocols serve the credit markets well — yet each encounters structural limitations when applied to true block-sized execution.

Portfolio trading has proven effective for risk mitigation and transference across baskets of bonds. However, an electronic solution for single-name blocks — large-notional orders where information sensitivity, adverse selection, and liquidity formation are paramount — largely remains unresolved. Each protocol below is well-suited to the use case it was designed for. The challenge is that none was architected to deliver the combination of anonymity, multilateral price formation, and conditional size protection that block execution requires.

1

Request for Quote (RFQ)

The dominant protocol for credit — with inherent size constraints

RFQ is the most widely adopted electronic protocol in credit markets and serves the majority of flow well. At block size, however, its bilateral structure introduces limitations:

  • Single-dealer balance sheet: each quote reflects only one dealer's risk capacity.
  • Defensive pricing at size: spreads widen non-linearly as notional grows.
  • Information leakage scales with outreach: additional RFQs reveal size and urgency.
  • No multilateral size formation: RFQ cannot clear a block across many participants simultaneously.

Current venues: Bloomberg FIT/BOLT, Bonds.com, Dealerweb, MarketAxess, Tradeweb, Trumid

“RFQ is the bread-and-butter market structure for credit because the client shows their hand first … and then the dealer comes back with pricing.”— Head of Credit Trading at a major RFQ platform provider

Net effect: At block size, execution becomes a sequential bilateral negotiation — concentrating counterparty risk and limiting price competition.

2

Central Limit Order Book (CLOB)

Proven in equities and rates — structurally challenged for bond blocks

CLOBs provide transparent, continuous price discovery and have transformed equities and on-the-run Treasuries. When applied to large corporate bond orders, however, the architecture works against the block trader:

  • Price-time priority: large orders must 'walk' through multiple price levels, incurring market impact.
  • Displayed liquidity creates signaling risk: visible size and side reveals intent, causing liquidity to fade.
  • No conditional size protection: the book cannot express 'all-or-nothing' block intent.
  • Time exposure compounds impact: each partial fill worsens the execution price of remaining quantity.

Current venues: Bonds.com, BrokerTec, Dealerweb, Fenics, OpenYield, Tradeweb

Net effect: A single economic block is decomposed into many mechanical trades — increasing information leakage and total execution cost.

3

All-to-All Crossing Networks

Broadened access to liquidity — dependent on coincidence for blocks

All-to-all platforms have meaningfully expanded the pool of counterparties and increased competition in credit markets. For block-sized orders, the crossing model faces structural constraints:

  • Coincidence requirement: blocks execute only if both sides independently match CUSIP, size, price, and timing.
  • No active price formation: slight price differences prevent clearing; near misses fail silently.
  • Instrument fragmentation: thousands of non-fungible bonds disperse latent interest across a vast security universe.
  • No conditional size commitment: cannot express "trade only if $50mm clears at one price."
  • Participation asymmetry: buy-side consistently seeks liquidity, while sell-side risk capital is episodic.

Current venues: Bonds.com, LiquidityEdge, Liquidnet (FI), MarketAxess, Tradeweb

Net effect: Block liquidity appears sporadically and unpredictably — execution is possible but not reliably constructible.

4

Streaming Protocols

Exceptional efficiency for flow — capacity-constrained at block size

Streaming has dramatically improved execution speed and consistency for standard-sized orders, and now represents a significant share of electronic credit volume. At block notionals, however, the model encounters limits:

  • Single-dealer balance sheet constraint: each streamed price reflects one dealer's inventory, capital, and risk appetite.
  • Liquidity thins at size: as notional increases, spreads widen or streams are withdrawn entirely.
  • No multilateral competition: dealers stream independently; liquidity is not aggregated into a single clearing price.
  • No conditional size expression: cannot express "trade only if $50mm clears at one price."
  • Adverse selection risk: dealers pull size when information leaks or asymmetric flow is suspected.

Current venues: Bloomberg FIT/BOLT, Bonds.com, BrokerTec, Dealerweb, Fenics, LiquidityEdge, Tradeweb, Trumid

Net effect: Streaming excels at rapid execution for standard clips — but is not designed to aggregate liquidity at block scale.

A Purpose-Built Approach
5

MatrixCross — The Fifth Protocol

Anonymous · Automated · Built for Size

MatrixCross was purpose-built for the use case the four protocols above were not designed to address: anonymous, multilateral execution of large-notional bond orders in seconds.

Protocol characteristics

  • Complete anonymity — identity, trade side, and reserve price are never disclosed to counterparties.
  • Multilateral price formation — the full electronic participant base competes to establish a clearing price and size.
  • Conditional size protection — blocks clear at one price or not at all; no partial fills, no residual exposure.
  • Execution in seconds — what typically requires 12–15 minutes of bilateral outreach completes electronically.

Structural positioning

  • Where bilateral protocols limit counterparty competition, MatrixCross aggregates liquidity across all participants.
  • Where crossing networks depend on coincidence, MatrixCross constructs actionable liquidity through its pricing matrix.
  • Where displayed protocols risk information leakage, MatrixCross protects every element of the order that could create adverse selection.

Result: Optimized execution outcomes with minimized information risk — purpose-built electronic infrastructure for the block segment of fixed income.

We welcome a conversation with anyone interested in learning more about MatrixCross.

Get in Touch